No, they didn't. What Samsung did was release software that mimicked facial recognition. In fact, it mimicked it so well, people were convinced that Samsung did it first. What Samsung actually did was implement software that would sometimes coincidentally "recognize" a face, and more often would "recognize" a photo of the target user. What Samsung didn't do was implement facial recognition that was reliable and secure, which I think is what most people think of when they think "facial recognition".
This raises a major question though - what's the use case and threat model for facial recognition? If you're remotely concerned about security, clearly you're not going to use it. If you only care about preventing the casual attacker, either option seems good enough.
I just don't see a security model in which Apple has substantially improved it. The security of it is terrible in either case - all Apple did was make it more expensive.
> What Samsung actually did was implement software that would sometimes coincidentally "recognize" a face, and more often would "recognize" a photo of the target user
No way unique to Samsung. Every implementation I've seen fails at this, even the latest one in Windows 10.
If you care about security, don't use it. That's my advice.
No, they didn't. What Samsung did was release software that mimicked facial recognition. In fact, it mimicked it so well, people were convinced that Samsung did it first. What Samsung actually did was implement software that would sometimes coincidentally "recognize" a face, and more often would "recognize" a photo of the target user. What Samsung didn't do was implement facial recognition that was reliable and secure, which I think is what most people think of when they think "facial recognition".